• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Spatial Audio -- warning from a producer/engineer

Chrispy

Master Contributor
Forum Donor
Joined
Feb 7, 2020
Messages
7,751
Likes
5,910
Location
PNW
So "remixing" for Atmos and MQA are about the same level of expertise/commitment to the final product? At least Atmos has a multich effect? Don't see buying into bullshit like mqa myself and wondering why I'd want to for "Atmos". If it's a good multich mix great, if it's a matrix applied with little thought...meh.
 
OP
A

Ata

Senior Member
Forum Donor
Joined
May 6, 2021
Messages
388
Likes
333
Location
Adelaide, Australia
You said in post #22 of this thread:

MFSL reissues of Pink Floyd were definitely a step forward for the albums involved, thanks to Alan Parsons involvement.
I rest my case. ;-)

No, my statement is not the same as:

Well, you did argue that the new 3D mixes were a 'step up' from other mixes partly because (some of) the original personnel were involved.

because:
- I made no argument that all "new mixes" are a step up. My argument was that some can be;
- I made no reference to 3D as I have no experience let alone a formed opinion.

More importantly, I enjoyed a lot of Can last night!
 

dfuller

Major Contributor
Joined
Apr 26, 2020
Messages
3,335
Likes
5,050
The up-scaling from stereo masters is... Not great. It sounds washed out and weird. This won't sound good until actual Atmos masters are done, and for the life of me I can't be convinced of the benefit for headphone users. I dunno about you but I wouldn't want the music to shift in headphones when I turn my head, that sounds awful.
 

rdenney

Major Contributor
Forum Donor
Joined
Dec 30, 2020
Messages
2,235
Likes
3,856
That Marvin Gaye demo is just horrible. I don't know where to start. Sounds like yet another frequency based chopping up along with heavy attack filtering and a dull sound, almost like a Dolby processing on a non-dolby analog tape from years gone by. I mean, does anyone think that is an improvment??

And let's play the spatial demos a bit louder and tell the sheep all the wonderful things they are hearing. :facepalm:

All the more reason to stay away from anything other than the original recordings, preferably on CD as your own personal 'masters'.
I look fondly at my obsolete library of physical CDs, and my desire to expand same with ebay purchases and the like. The streaming services are too driven by priorities I do not share, or that do not fit my use cases or objectives.

Rick "who doesn't stream" Denney
 

rdenney

Major Contributor
Forum Donor
Joined
Dec 30, 2020
Messages
2,235
Likes
3,856
Define “stereo.”

How would you characterize, e.g. the Reiner/CSO performances recorded in “stereo” as the format was intended to be (3-channel LCR)before it had to bow to the reality of the day’s primitive distribution technology (crammed down to 2 channel).
But they were mixed in stereo.

If the original three-channel recording was made available for use in appropriately configured 3-channel systems, then nobody has a problem with that (as long as we have the option as buyers).

But that's not what I see happening here--I see stuff that was recorded with multiple mikes on a large number of tracks, mixed to stereo, and then the stereo mixed is processed to simulate multiple tracks.

Or the original tracks are remixed for the new format without input from those who made the original mix decisions.

Given that most multitrack recordings created the stereo image using pan pots on a mixer board, I would think figuring out how to remix that in any kind of multichannel presentation would require a lot of care. Just as with a lot of assembly-line digital conversions of historical analog recordings that contain unlistenable artifacts (example: I have a CD made from a Harry James recording that needs a low-pass filter set to about 9K to be listenable at all--there are all sorts of audible mixing products), assembly-line conversions to something like Atmos will cause head-shaking by future generations.

Rick "keeping his old CD's and even vinyl records" Denney
 

Ralph_Cramden

Major Contributor
Joined
Dec 6, 2020
Messages
2,574
Likes
3,468
Of course they knew what black and white did to the picture and used it. Simpler times doesnt mean simpler people.

Indeed. Imagine a great Film Noir like The Third Man in color! Simply wouldn’t work.
 

Billy Budapest

Major Contributor
Forum Donor
Joined
Oct 11, 2019
Messages
1,810
Likes
2,674
So, I’ve sampled most of the 100 or so Atmos demo tracks on Apple Music.

Some do not sound noticeably different than their stereo counterparts.

Many others sound like they have different balance between the instruments and different EQ, but they do not sound more enveloping.

A small minority of the tracks do, in fact, have sound effects that appear somewhat hazily to come from behind or above your head.

I would not describe the effect as “uncanny” in realism, as some have.

The only tracks that I have found to be engaging are ones that are primarily vocal, with sparse instrumentation. Norah Jones’ first album (which although she has a pretty voice, I find boring) is actually improved by the effect.

I do not think the problem is the Atmos remixing. Atmos sounds fine in multi speaker surround sound systems. The problem is the virtual surround algorithms used by Apple to simulate surround sound in headphones. They just don’t work as well as they are being touted.

To sum up: “spatial audio” does not sound like the revolution in audio that Apple marketing folks are making it out to be. Perhaps they can improve the technology. At this point, although the “spatial audio” tracks sound different than the stereo versions, they don’t create a convincing surround experience.
 

NiagaraPete

Major Contributor
Forum Donor
Joined
Jun 23, 2021
Messages
2,185
Likes
1,953
Location
Canada
Listening to a wonderfully produced Porcupine Tree at the moment.

ah the thread. I remember when CD’s were pushed out. Horrible in most every way. I quit the business shortly thereafter.
 

Cbdb2

Major Contributor
Joined
Sep 8, 2019
Messages
1,530
Likes
1,485
Location
Vancouver
Multichannel audio proccesed from a stereo mix has little hope of sounding great, its like stereo from mono. Multichannel mixed from the multitrack recording can easily sound better than a stereo mix. The problem is that the mix from multitrack is half the work in creating the final song. It involves, for each of the 24 tracks, setting levels, EQ, compression, probably some delay, reverb, maybe some chorusing, flanging etc. And then different mix desks,compressors (etc) sound a little different.

Now, some dumb ass at apple asks you to remix Dark Side of the Moon from the multitrack. How close do you think the tone and ballance of the instruments and voices will be to the original stereo mix? So Ild bet the first option is what there doing. The one that doesnt work properly.
 

NiagaraPete

Major Contributor
Forum Donor
Joined
Jun 23, 2021
Messages
2,185
Likes
1,953
Location
Canada
Listening to a wonderfully produced Porcupine Tree at the moment.

ah the thread. I remember when CD’s were pushed out. Horrible in most every way. I quit the business shortly thereafter.
I have to amend my post with an attachment from the archives.
 

Attachments

  • 3F997CA7-52E3-4C99-8A2E-330E8755284A.jpeg
    3F997CA7-52E3-4C99-8A2E-330E8755284A.jpeg
    247.3 KB · Views: 88

Billy Budapest

Major Contributor
Forum Donor
Joined
Oct 11, 2019
Messages
1,810
Likes
2,674
The stereo version of that old Marvin Gaye tune sounded to me like distortion, compression and even some reverb had all been added. The Atmos version sounded better to me, only because it sounded cleaner. I think the demo is rigged. The Atmos version sounds better, but not because of Atmos. It sounds better because it's been cleaned up whereas the stereo version comes from some completely different copy that hasn't been cleaned up, and likely has been distorted in various ways. This kind of demo is almost guaranteed to be rigged. Why wouldn't they do this kind of thing? Is anyone watching over their shoulders to make certain that they created the Atmos version from the stereo version and that the Atmos version wasn't given any special treatment as compared to the routine Atmos conversion of other material?

By the way, I couldn't stand listening to that narrator guy, or whatever he's referred to. His voice sounded whiny and wimpy, and he ran words togetherlikeeverythingwasallonelongword. Why would they do something like this and not hire someone with a good speaking voice?

I haven't much liked anything that Apple has done in a very long time. In fact the only thing they've ever done that I much liked was when they ditched their old operating system and adopted a variant of UNIX. That was a great move, but there still way too much stuff in Mac OS that is carried over from that ridiculous softcontraption that Jobs first introduced back in the mid-early eighties. And as far as I'm concerned most all of their application software, including iTunes, is grotesque.

I do not believe in synthesizing extra tracks from a two-track recording. A regular stereo recording made years ago does not contain any information that corresponds to an additional, reflected-off-the-ceiling track. You cannot create something from nothing. If the process by which the off-the-ceiling tracks are being created is any different from what could otherwise be synthesized within the playback equipment, then it is necessarily a creative process. To me, it isn't much different from someone taking an old original painting and painting over it with their own personal touches, to make it look more the way they think it should look. I don't really see any difference. But it is very much the sort of thing that Apple would do, in cahoots with some other company. (Shhhh ... don't anyone mention Beats.) It makes me sad to think that all those sheep-people who are perpetually crowded into the Apple store at the mall are now going to start listening to music that has been modified, no longer the same as the original, and they are going to think of it as an improvement, because some of it is intentionally reflected off the ceiling. Very, very sad. Somehow, the idea of people preferring to listen to old-style amplifiers that add distortion that they think equates to "better" sound quality doesn't seem as bad as did up until several minutes ago. I think maybe I'm going to patent a process that will detect the main tempo of recorded music and add cowbell. Instead of it being a preset amount, it will have a control to let the listener add cowbell to suit their personal taste.
I don’t think the spatial audio demo is “rigged.” Although Apple pursues the almighty dollar like nobody’s business, they are a very ethical company. It would go against their core values to take the stereo track and add distortion to it in an effort to hoodwink listeners.

That all being said, I think the stereo version of the track sounds better than the “spatial audio” version, not worse.

I don’t think the Atmos mixing is to blame. Atmos (and other object-based surround formats) can sound, well, awesome on a multichannel, multi-speaker surround system. Rather, whatever psychoacoustic processing Apple is using to attempt to synthesize multiple channels by using just two headphone drivers is to blame. It just isn’t blow-me-away good. It’s just kind of meh.
 

DavidMcRoy

Addicted to Fun and Learning
Joined
Jul 10, 2018
Messages
574
Likes
989
Multichannel audio proccesed from a stereo mix has little hope of sounding great, its like stereo from mono. Multichannel mixed from the multitrack recording can easily sound better than a stereo mix. The problem is that the mix from multitrack is half the work in creating the final song. It involves, for each of the 24 tracks, setting levels, EQ, compression, probably some delay, reverb, maybe some chorusing, flanging etc. And then different mix desks,compressors (etc) sound a little different.

Now, some dumb ass at apple asks you to remix Dark Side of the Moon from the multitrack. How close do you think the tone and ballance of the instruments and voices will be to the original stereo mix? So Ild bet the first option is what there doing. The one that doesnt work properly.

I believe that the best course of action to take in preparing a release of a Dolby Atmos version of a legacy 2-channel stereo recording would be to create a whole new mix from the original multitrack studio recordings. It would probably take a proverbial “act of Congress” to make that happen, but that would be ideal.

Based on my own personal, day-to-day experience with a 7.4.4 Dolby Atmos speaker system (the middle “4” represents my four independently DSPed subwoofers,) is that for an enthusiast of immersive sound like me, the Dolby Audio Surround upmixer that’s built into my Dolby Audio processor does a ridiculously credible job on 2-channel movies, TV and music, as well as on native 5.1 material. It simply has to be experienced to be fully appreciated, being mercifully light-handed most of the time on music. Dolby Labs really did their homework on that algorithm. It’s pretty hard to fault.

When it comes to a completely different animal, binaural headphone ”Dolby Atmos” remixes, I don’t quite know what to make of most of what I’ve heard so far. The decisions made by the Dolby Renderer that’s at the heart of the process, and/or the mixing engineers using it are all over the map right now, leading me to reserve judgment until more engineers have more experience and interaction with Dolby Labs than seems to have happened to date.

I view Dolby Atmos in it’s various incarnations as just a newish “tool in the toolbox” for audio professionals to master. I think it’s clearly “a way forward.”
 
Last edited:

KeithPhantom

Addicted to Fun and Learning
Forum Donor
Joined
May 8, 2020
Messages
641
Likes
658
I don’t think the spatial audio demo is “rigged.” Although Apple pursues the almighty dollar like nobody’s business, they are a very ethical company. It would go against their core values to take the stereo track and add distortion to it in an effort to hoodwink listeners.
I looked into the source of this claim, and only they seem to be claiming this kind of things. I am skeptical of the source and cannot really take it at face value. I would need evidence that this is true in order to believe it.
 

Haint

Senior Member
Joined
Jan 26, 2020
Messages
346
Likes
452
From a practical point of view, if they depended solely on the original artist, producers, or mixer to do these, there would only be like 5-10 songs available, you'd be lucky to get another handful every calendar year, and each remastered album would probably cost several million dollars (major artists and producers simply aren't going to waste their time for anything less).

On a related note, has anyone compiled a list of the particularly standout and quality Atmos tracks?
 
Last edited:

DavidMcRoy

Addicted to Fun and Learning
Joined
Jul 10, 2018
Messages
574
Likes
989
I looked into the source of this claim, and only they seem to be claiming this kind of things. I am skeptical of the source and cannot really take it at face value. I would need evidence that this is true in order to believe it.

Re: was the Marvin Gaye Dolby Atmos demo “rigged,” well they definitely added stereo (!) reverb to the MONO cut.
 

Studio Guy

Member
Joined
Apr 28, 2021
Messages
31
Likes
30
Sorry but at the end of the day that is one man's opinion along with an appeal to authority.
I think people are confusing bad mixes with a limitation of the stereo format. The format is not limited. If the desire is to have the audio image sound "bigger" without making them sound far away, a good engineer can easily do that in stereo. Putting the reverb behind your head is not going to help a bad mix.
 

Billy Budapest

Major Contributor
Forum Donor
Joined
Oct 11, 2019
Messages
1,810
Likes
2,674
Re: was the Marvin Gaye Dolby Atmos demo “rigged,” well they definitely added stereo (!) reverb to the MONO cut.
The “mono” cut they used was obviously just a folded-down version of the stereo cut. I don’t think that a true mono mix of What’s Going On is even available on Apple Music. However, the question is whether Apple took the stereo mix of What’s Going On and handicapped it by adding distortion, as one person claimed. That is highly doubtful, not to mention unethical and potentially illegal (if the demonstration track is construed as an advertisement for Apple Music spatial audio tracks).
 
OP
A

Ata

Senior Member
Forum Donor
Joined
May 6, 2021
Messages
388
Likes
333
Location
Adelaide, Australia
So, I’ve sampled most of the 100 or so Atmos demo tracks on Apple Music.

Some do not sound noticeably different than their stereo counterparts.

Many others sound like they have different balance between the instruments and different EQ, but they do not sound more enveloping.

A small minority of the tracks do, in fact, have sound effects that appear somewhat hazily to come from behind or above your head.

I would not describe the effect as “uncanny” in realism, as some have.

The only tracks that I have found to be engaging are ones that are primarily vocal, with sparse instrumentation. Norah Jones’ first album (which although she has a pretty voice, I find boring) is actually improved by the effect.

I do not think the problem is the Atmos remixing. Atmos sounds fine in multi speaker surround sound systems. The problem is the virtual surround algorithms used by Apple to simulate surround sound in headphones. They just don’t work as well as they are being touted.

To sum up: “spatial audio” does not sound like the revolution in audio that Apple marketing folks are making it out to be. Perhaps they can improve the technology. At this point, although the “spatial audio” tracks sound different than the stereo versions, they don’t create a convincing surround experience.

Did you listen exclusively on headphones, and which ones?

I have Airpods Pro and listened to few tracks from the demo areas, I concur they are underwhelming. I did not get spatial audio, though -- turning my head changes nothing. Watching Netflix with an Atmos soundtrack, however, sounds great, and spatial audio works well. My guess is that they are still ironing out some kinks in the integration. Netflix Atmos sounds very very good on these otherwise mediocre headphones, while Atmos music is just... mediocre.

I also listened to Atmos tracks on my modest 5.1 sound system (soon to be 5.2.2), and there it sounds much better than on the Airpods Pro.

Could you share a list of albums that sounded best to you? Thanks!
 

Billy Budapest

Major Contributor
Forum Donor
Joined
Oct 11, 2019
Messages
1,810
Likes
2,674
Did you listen exclusively on headphones, and which ones?

I have Airpods Pro and listened to few tracks from the demo areas, I concur they are underwhelming. I did not get spatial audio, though -- turning my head changes nothing. Watching Netflix with an Atmos soundtrack, however, sounds great, and spatial audio works well. My guess is that they are still ironing out some kinks in the integration. Netflix Atmos sounds very very good on these otherwise mediocre headphones, while Atmos music is just... mediocre.

I also listened to Atmos tracks on my modest 5.1 sound system (soon to be 5.2.2), and there it sounds much better than on the Airpods Pro.

Could you share a list of albums that sounded best to you? Thanks!
I listened on AirPods Pro. I have listened to Atmos tracks on my regular Atmos system before but did not do so with the Apple Music tracks as they are touting “spatial audio” as a headphone technology—and specifically, a technology geared towards headphones with the H1 processor.

I listened to the 100 (or so) song playlist, one of a few playlists on Apple Music showcasing Atmos and spatial audio. I’ll re-listen and note what I think are the best tracks and then post here.
 
  • Like
Reactions: Ata
Top Bottom